Urinary tract infections (UTIs) in women begin with the attachment of infecting microorganisms to vaginal and subsequently to bladder epithelium. In our prior studies, we utilized primary bladder epithelial cell (BEC) cultures as a model to define the role and regulation of glycosphingolipids (GSLs) as attachment sites for uropathogenic E. coli, later extending these studies to a new model of primary vaginal epithelial cell (VEC) cultures. GSLs do not occur randomly in cell membranes, but instead are organized into specialized membrane domains such as caveolae, characterized by enrichment in cholesterol, sphingomyelin, glycolipids such as ganglioside GM1, lipid-anchored proteins, and caveolin. These domains are involved in a wide variety of key cellular functions, including transport of cholesterol, macromolecular solute transport, tumor suppression, and signal transduction. Recent evidence demonstrates that caveolae are key molecules in the initial host response to attaching uropathogenic E. coli in mast cells, mediating the uptake of organisms. We have preliminary data demonstrating that caveolin-1 is present in primary cultured VEC and BEC monolayers and native vaginal tissue sections, and that native vaginal epithelial cells contain GM1, a key GSL of caveolae in mast cells and other tissues. The central hypothesis of this proposal is that caveolae occur in bladder and vaginal epithelium and participate in the initial epithelial responses to attachment of uropathogenic E. coli. We will pursue the following specific aims: (I) To conclusively demonstrate that caveolae occur in bladder and vaginal epithelium, we will structurally and biochemically characterize these structures in primary cultured BEC and VEC, including defining GSLs localized to caveolae in these cells; (2) To address the hypothesis that caveolae contain key receptor molecules for uropathogenic E. coli, we will determine if caveolin co-localizes with the globoseries GSLs in primary cultured BEC and VEC and/or with mannosylated glycoproteins that bind Type 1 fimbriated E. coli; (3) To investigate the hypothesis that caveolae mediate uptake of Type 1 fimbriated E. coli into urogenital epithelium, we will investigate invasion of cultured primary human BEC by this organism, testing for co- localization of engulfed bacteria with caveolar markers and investigating if disruption of caveolae prevents bacterial uptake; and (4) using established neo-organ models of bladder and vaginal epithelium, we will demonstrate key findings from the monolayer systems, such as the presence of caveolae, bacterial invasion, and mediation of this process by caveolae. These studies will advance knowledge of the role of GSLs and caveolae, fundamental cellular components, in the pathogenesis ofE. coli UTI.